PCOS is among the most common disorders of adolescent and premenopausal women, affecting approximately 710% of this population. It is a high priority and overarching women's health problem with substantial reproductive and metabolic morbidities throughout the lifespan. Dunaif's recent studies on the mechanisms of insulin resistance in PCOS have revealed the surprising finding that defects in skeletal muscle insulin action are acquired secondary to a factor (or factors) in the in vivo environment (Project 1). Dunaif and colleagues' family studies have shown that hyperandrogenernia is the major reproductive phenotype in PCOS kindreds (Figure 3). Urbanek and colleagues have compelling evidence that this phenotype is linked with a marker, D19S884, on chromosome 19p in the region of the insulin receptor gene (Project 2). This marker is also associated with a metabolic phenotype in PCOS women as well as in their brothers characterized by decreased insulin secretion, particularly in response to sulfonylurea (Project 1). Abbott and colleagues have shown that many of the phenotypic features of PCOS, such as ovarian hyperandrogenism, polycystic ovaries, increased LH levels, anovulation, central adiposity and decreased insulin secretion can be produced in rhesus monkeys by intrauterine testosterone exposure (Project 3). Levine has obtained evidence that one mechanism for some of these androgen actions is decreased function of ATP-sensitive potassium channels (K+ATP channel) in gonadotropin releasing hormone (GnRH) containing neurons and in pancreatic islet P-cells (Project 4). Sulfonylureas stimulate insulin secretion through activation of one of these channels, known as the sulfonylurea receptor, and the same channel complex appears to function in GnRH neurons. These observations have led to a paradigm shift in our concept of the pathogenesis of PCOS. Exposure of the fetus to androgens could result in the reproductive phenotype and the pancreatic P-cell dysfunction characteristic of PCOS. We propose to test the hypothesis that hyperandrogenernia resulting from variation in a gene in linkage disequilibrium with D I 9S884 causes many of the phenotypic features of PCOS by prenatal androgen programming. This hypothesis will be directly tested in two animal models and in translational human studies. The metabolic phenotype associated with the chromosome 19p PCOS susceptibility gene will be defined and this susceptibility gene will be identified. These studies will elucidate the pathogenesis of PCOS and provide the potential for molecular diagnosis of the syndrome. These objectives will be accomplished in four highly synergistic and interactive research projects.